Differences in Obesity Management Among Physicians

Abstract

Despite the strong recommendations of guidelines, intensive obesity management is not offered to all obese patients. This study aimed to examine differences in obesity management between primary care physicians (PCPs) and non-PCPs. A cross-sectional study was performed using the 2006–2007 National Ambulatory Medical Care Survey. Adults (age ≥20 years) with obesity (body mass index (BMI)≥30 kg/m² or obesity diagnosis using International Classification of Diseases, Ninth Revision, Clinical Modification code 278) were included in the study cohort. A multivariate logistic regression model was constructed to examine differences between PCPs and non-PCPs (primary independent variable) for obesity management (dependent variable) while controlling for predisposing, enabling, and need characteristics per Anderson's behavioral model. In all, 32.66% of 214 million visits by obese patients in 2006–2007 resulted in obesity management. PCPs were 2.38 times more likely to provide obesity management compared to non-PCPs (odds ratio [OR]=2.37; 95% confidence interval [CI]: 1.69, 3.36). Patients who had preventive visits (OR=2.23; 95% CI: 1.50, 3.32) and chronic visits (OR=1.93; 95% CI: 1.46, 2.55) were more likely to receive obesity management than patients who had acute visits. More time spent with physician, more comorbid conditions, and BMI ≥ 40 significantly increased the likelihood of receiving obesity management, while older age and smoking reduced the likelihood of receiving obesity management. Only one third of ambulatory care visits in 2006–2007 resulted in obesity management. A difference in obesity management was noted between PCPs and non-PCPs. Future research should aim to identify the reasons for these observed differences, ensure equitable access, and address the undertreatment of obesity. (Population Health Management 2012;15:287–292)

Introduction

O besity is one of the major causes of morbidity and mortality, and is a growing public health concern, especially in developed countries.^1,2 With a current prevalence of 33% for obesity (body mass index [BMI] ≥30 kg/m²) and 68% for overweight (BMI≥25 kg/m²), the United States ranks number 1 among all developed countries for overweight/obesity prevalence.^3,4 Obesity causes or increases the risk for several comorbidities such as type 2 diabetes, hypertension, and cardiovascular diseases.⁵ It also is reported that obesity raises an individual's annual medical care cost by $2826.⁶ Ample evidence exists in the literature to demonstrate that even a moderate reduction in weight may have significant benefits in terms of reducing the risk of morbidity and mortality, yielding subsequent cost savings.^7
–9

According to the National Heart, Lung, and Blood Institute guidelines, behavioral therapy (ie, diet, exercise) is the first line of therapy for obesity management; if behavioral therapy is not effective in reducing weight, pharmacotherapy should be initiated.¹⁰ The US Preventive Services Task Force (USPSTF) strongly recommends that clinicians should screen all patients for obesity and offer intensive counseling and behavioral interventions to promote sustained weight loss.¹¹ Nonetheless, the obesity management rate has remained low in the United States. Two studies using the Behavioral Risk Factor Surveillance System estimated that the obesity counseling rate was 40% in 2000.^12,13 The obesity counseling rate was 24% in 2003/2004 and less than 20% in 2005 in ambulatory care settings. Existing evidence confirms that physician-offered obesity counseling has a significant impact on motivating patients to engage in weight loss activities and to achieve greater weight loss.^14
–16

Despite consensus that obesity is a disease in need of medical attention, many obese patients fail to receive the required medical care.¹⁷ Previous literature reports the underdiagnosis of obesity and infrequent counseling provided on lifestyle modification in ambulatory care settings across the United States.^18,19 A recent study found that most obese patients did not receive obesity counseling in 2005 and also documented differences between physician specialties for weight reduction, diet, and exercise counseling.²⁰

The objectives of the current study were to examine differences in obesity management between primary care physicians (PCPs) and non-PCPs and to provide a comprehensive picture of current obesity management practices using recent available data sets, while considering any kind of weight-related counseling or prescriptions for anti-obesity medications.

Methods

Data source

This study used National Ambulatory Medical Care Survey (NAMCS) data from 2006 and 2007. The NAMCS is a national probability survey of patient visits to office-based physicians that is conducted annually by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention. It utilizes a multistage probability sampling design. The first stage sample included 112 primary sampling units (PSUs), defined as a county or group of counties, a county equivalent (parish or independent city), a town or township, or a metropolitan statistical area. The second stage included a probability sample of practicing physicians selected from the American Medical Association and the American Osteopathic Association master file. In the third stage, patient visits within the annual patient panels of the practices of sample physician were selected randomly. NAMCS data have been used widely by many researchers to examine health care services use in ambulatory care.^19,20

Physicians completed a patient record form for each patient visit by providing information regarding patient demographics, BMI, the reason for the visit, physician diagnosis, types of counseling provided, prescription medication use, and physician characteristics. In an effort to improve the quality of the data collected, all patient record forms with differences between coders or illegible entries for the reason for the visit, the diagnosis, procedures, and medication items were reviewed and adjudicated at NCHS. For nonmedical coding items, the error rate ranged from 0.3% to 0.4%, whereas the medical coding error rate ranged from 0% to 1.6%. The NAMCS uses patient visit weight, which enables researchers to extrapolate results at the national level. The patient visit weight accounted for multistage selection probability, nonresponse adjustment, and other adjustments to reflect the universe of office-based patient visits in the United States.

Study design

A retrospective cross-sectional study design was utilized that included all adults (age ≥20 years) who were obese. Obesity was defined as having a BMI ≥30 kg/m² or a diagnosis of obesity (International Classification of Diseases, Ninth Revision, Clinical Modification code 278).

Conceptual framework and variables

This study used Anderson's behavioral model of health service use. The model includes predisposing, enabling, and need characteristics of an individual, which in turn determine obesity management.²¹

The outcome variable of interest was obesity management, which was defined as either an anti-obesity medication prescription or counseling received for obesity. Obesity counseling included diet/nutrition, exercise, or weight reduction counseling offered to the patient. In NAMCS, the patient report form asks the physician if he or she is the patient's PCP. Based on the response, physicians were divided into 2 categories: PCP and non-PCP.

Predisposing characteristics included age, race/ethnicity (non-Hispanic whites, non-Hispanic blacks, Hispanics), and sex. Enabling characteristics refer to community and personal resources and include insurance (private, public, no insurance), metropolitan statistical area, region, amount of time spent with physician, and tobacco use. Need characteristics refer to the patient's perceived and evaluated need to seek health care services. These include BMI, the reason for the visit, and risk status based on comorbidities. There were 3 categories of BMI: class I (BMI ≥30 and <35), class II (BMI ≥35 and <40), class III (BMI ≥ 40). The reason for the visit was categorized as acute, chronic, or preventive. Risk status was divided into 2 categories: very high/high versus low/no. The very high/high risk category included coronary artery disease, diabetes mellitus, sleep apnea, osteoarthritis, gallstones, stress incontinence, hyperlipidemia, cigarette smoking, hypertension, men aged ≥45 years, and women aged ≥55 years; the low/no risk category included the remaining obese adults.

Statistical analysis

All statistical analyses were adjusted for NAMCS complex survey design in order to extrapolate findings at the national level. Descriptive statistics were used to describe the study cohort. Chi-square test and the Student t test were performed to examine for differences between obese adults who received obesity management and those who did not for categorical and continuous variables, respectively. A multivariate logistic regression model was constructed to examine physician differences in obesity management controlling for covariates, predisposing, enabling, and need characteristics, per Anderson's behavioral model.

All statistical analyses were performed using SAS 9.2 (SAS Institute Inc, Cary, NC). Statistical significance was considered at P<0.05. The study was approved by the University of Houston Committees for the Protection of Human Subjects as exempt.

Results

Obese patients made 214,430,000 (95% confidence interval [CI]: 187,807,000 to 241,053,000) ambulatory care visits in 2006 and 2007. Of those visits, one third (32.66%) resulted in obesity management. Patients who received obesity management were more likely to be older, female, and of non-Hispanic white race. Table 1 reports the characteristics of study population.

Table 1.

Characteristics of the Study Population (Obese Adults)

	Total number of obese patients (Number in thousands)	Received obesity management (%)	Did not receive obesity management (%)	P value
All adults	214,430	70,030 (32.66%)	144,400 (67.34%)
Sex
Female	129,386	61.22	59.91	0.541
Male	85,044	38.78	40.08
Race/Ethnicity
Non-Hispanic Blacks	28,878	12.88	13.75	0.012
Hispanics	26,542	15.57	10.83
Others	7,729	4.48	3.18
Non-Hispanic whites	151,281	67.07	72.24
Age (Mean±SD)	44.60±0.65	52.52±0.44	43.59±0.71	<0.0001
Region
Other	125,179	56.47	59.30	0.470
South	89,251	43.53	40.70
Urban residence
MSA	185,841	87.82	86.11	0.349
Non-MSA	28,589	12.18	13.89
Insurance status^*
Private	118,313	59.29	59.17	0.046
Public	67,677	31.48	35.04
Uninsured	13,825	9.23	5.78
Reason for visit^*
Preventive	33,952	18.91	14.71	<0.0001
Chronic	108,977	57.13	48.94
Acute	68,256	23.96	36.35
Tobacco use
Smoker	28,038	15.15	20.42	0.0006
Nonsmoker	122,575	84.85	79.58
Time spent with physician
>24 minutes	60,063	32.03	26.06	0.0001
13–24 minutes	113,105	53.72	52.27
0–12 minutes	41,262	14.25	21.67
Comorbidity
Very high/high	148,944	70.00	69.20	<0.7051
Low/no	65,486	30.00	30.80
BMI^*
>40	44,189	30.17	17.15	<0.0001
>35 to ≥40	53,464	24.28	26.39
>30 to ≥35	110,182	45.56	56.46
Provider
PCP	108,787	63.42	49.31	<0.0001
Specialist	92,926	36.58	50.69

Numbers do not add to 100% due to missing values.

BMI, body mass index; MSA, metropolitan statistical area; PCP, primary care physician.

Table 2 reports results for the multivariate logistic regression model. PCPs were more likely to provide obesity management compared to non-PCPs. Among the factors associated with an increased likelihood of receiving obesity management were spending a longer time with the physician, having very high/high comorbidities, having a class III BMI, and making a preventive or chronic disease visit. Older age and smoking reduced the likelihood of receiving obesity management. There was no difference in obesity management related to sex, race/ethnicity, region, urban residence, or insurance status.

Table 2.

Multivariate Logistic Regression Analysis for Differences in Obesity Management among Physicians

	Odds Ratio (95% CI)	P value
Provider
PCP	2.38 (1.69, 3.36)	<0.0001
Specialist	1.00 (Reference)
Sex
Female	0.97 (0.77, 1.21)	0.759
Male	1.00 (Reference)
Race/ethnicity
Non-Hispanic Blacks	0.76 (0.52, 1.10)	0.144
Hispanics	1.51 (0.92, 2.48)	0.103
Others	0.96 (0.51, 1.80)	0.891
Non-Hispanic whites	1.00 (Reference)
Age (increments of 1 year)	0.98 (0.97, 0.99)	<0.0001
Region
Other	0.78 (0.54, 1.14)	0.198
South	1.00 (Reference)
Urban residence
MSA	1.44 (0.97, 2.13)	0.070
Non MSA	1.00 (Reference)
Insurance status
Private	0.66 (0.34, 1.29)	0.227
Public	0.79 (0.41, 1.51)	0.474
Uninsured	1.00 (Reference)
Reason for visit
Preventive	2.23 (1.50, 3.32)	<0.0001
Chronic	1.93 (1.46, 2.55)	<0.0001
Acute	1.00 (Reference)
Tobacco use
Smoker	0.52 (0.39, 0.69)	<0.0001
Nonsmoker	1.00 (Reference)
Time spent with physician
>24 min	2.67 (1.81, 3.94)	0.002
13–24 min	1.89 (1.26, 2.82)	<0.0001
0–12 min	1.00 (Reference)
Comorbidity
Very high/high	1.46 (1.13, 1.89)	0.004
low/none	1.00 (Reference)
BMI
>40	1.88 (1.52, 2.34)	<0.0001
35–40	1.11 (0.90, 1.36)	0.328
30–35	1.00 (Reference)

BMI, body mass index; MSA, metropolitan statistical area; PCP, primary care physician.

Discussion

The study results highlight the undertreatment of obesity in the ambulatory care setting with only one third of obesity visits in 2006 and 2007 resulting in obesity management. The United States ranks number 1 in obesity prevalence among all developed countries. In an effort to reduce the disease burden and improve the nation's health, the USPSTF strongly recommends offering intensive counseling to each obese patient. In 2005, less than 20% of visits were provided obesity counseling.²⁰ In 2006–2007, this increased to 32.66%. However, given that only one third of the visits receive obesity management, the rate remains suboptimal.

Physicians play an influential role in educating patients about weight reduction. Patients who received weight reduction counseling from physicians had a significantly higher likelihood of trying to lose weight compared to their counterparts.¹² Physicians' weight reduction advice has a significant effect on patients' understanding of and motivation for weight loss.¹⁴ Studies also have shown that patients who received counseling were more likely to lose weight.^15,16

Several factors may explain why physicians may not provide obesity management to all patients, such as lack of time, patient noncompliance, lack of knowledge, and inadequate reimbursement.^14,22 Behavioral weight reduction counseling is not reimbursed by US health insurance companies, and they cover only a few pharmacological treatments.¹⁶ Moreover, to date, several anti-obesity drugs were withdrawn from the market, including the well-publicized withdrawal of fen-phen and the recent withdrawal of sibutramine. Physicians and patients may be reluctant to use anti-obesity medications because of the prospect of only modest weight loss and adverse effects. Nonetheless, behavioral therapies such as obesity counseling are devoid of adverse effects and previous studies have documented the effectiveness of obesity counseling in weight reduction.^15,16,23 Hence, physicians can help combat this growing epidemic by documenting and providing obesity management to all eligible patients.

Numerous factors (eg, clinician's life values, definitions of success, the availability of community resources, the patient's agenda and receptivity to the proposed counseling, the presence of teachable moments) play a role in the very complex decision by a physician to provide obesity management.^24,25 Previous studies have reported that the physician's belief that weight reduction advice would have little impact, that obesity is the responsibility of patients, and that obesity is hard to handle together may partly explain the observed differences in obesity management among physicians.^26,27 However, the current study did not control for the aforementioned factors as they were not recorded in the database.

Multiple studies have shown that obesity diagnoses are underdocumented.^18,19,28 Studies also have established a positive link between obesity diagnosis and obesity management.^28
–30 Thus, obesity management may be underdocumented. There may be other reasons why obesity was not included in diagnoses. For instance, the patient record form in NAMCS allows the physician to document only 3 diagnoses. If the patient had other major illnesses, the physician may not have had the opportunity to document obesity. Furthermore, as physicians are not reimbursed for providing obesity management, this lack of incentive may deter them from documenting the service. Including obesity as a diagnosis would have a positive impact on obesity management. One study compared obesity diagnosis and management rates in clinical settings before and after implementation of an electronic medical record (EMR). The results demonstrated a significant increase in obesity diagnoses (from 31% to 71%) and obesity treatment (from 39% to 59%).³¹ It is possible that the next generation of EMR data may increase documentation of obesity diagnoses and obesity management. Because EMR data capture information more comprehensively, future studies using EMR data may provide better insight into evaluating differences in obesity management among physicians.

Likelihood of obesity management was higher when patients made chronic visits and had high-risk comorbidities. As obesity can cause or increase the risk of several chronic diseases, this finding suggests that physicians make more of an effort to provide obesity management for this higher risk patient population. Furthermore, compared to acute care, visits for preventive care were more likely to end in obesity management because an acute visit usually is made to address a specific concern. Older age reduces the odds of receiving obesity management; this may be because the functional and/or physical limitations of older adults restrict them from performing vigorous exercise or dieting or because of the common belief that obesity may be harder to control at an older age. As the amount of time spent with the physician increased, patients were more likely to receive obesity management, which reflects increased interaction and discussion time to address their health problems in detail and gives the physician the opportunity to offer obesity management.

Obesity and smoking have been individually associated with the increased risk for several chronic conditions such as coronary heart disease, cerebrovascular disease, and cancer, and they adversely affect quality of life.^5,32
–34 Furthermore, several studies showed that obesity and smoking are independent risk factors for mortality but the combination of obesity and smoking poses higher mortality risks.^35

–38 This study identifies that obese smokers were significantly less likely to receive obesity management than nonsmokers. Future research should concentrate on this high-risk population to explore the reasons for this decreased rate as well as ways to address it.

The strengths of this study were that we used recent data from NAMCS 2006 and 2007, and that the study was guided by Anderson's behavioral model and provides a comprehensive view of obesity management by assessing any kind of intervention (ie, medical, behavioral) instead of an individual type of counseling. Nonetheless, the study had some limitations: counseling quality was not assessed because only yes/no responses were recorded on the form for weight reduction, diet, and exercise counseling. The database did not capture patients who have used nonprescription medications for weight loss. Even though there were some missing values for a few patients in the multivariate model in 1 or more of the explanatory variables, there was a large sample size from 2 years, and thus we believe this did not affect our results. Future research should focus on finding strategies to improve the obesity management rate offered in ambulatory settings.

Conclusion

Despite guideline recommendations, study findings suggest that only one third of obesity visits resulted in obesity management. PCPs were more likely to provide obesity management compared to non-PCPs, and obese smokers were less likely to receive obesity management compared to nonsmokers. Given that 1 in every 3 Americans is obese, proper documentation and management of obesity may help the nation combat this growing public health problem.

Footnotes

Disclosure Statement

Mr. Mehta, Mr. Patel, Mr. Parikh, and Dr. Abugosh disclosed no conflicts of interest.

References

Haslam

, James

. Obesity. Lancet, 2005; 366:1197–1209.

Ogden

, Yanovski

, Carroll

, Flegal

. The epidemiology of obesity. Gastroenterology, 2007; 132:2087–2102.

Flegal

, Carroll

, Ogden

, Curtin

. Prevalence and trends in obesity among US adults, 1999–2008. JAMA, 2010; 303:235–241.

Low

, Chin

, Deurenberg-Yap

. Review on epidemic of obesity. Ann Acad Med Singapore, 2009; 38:57–59.

Kopelman

. Health risks associated with overweight and obesity. Obes Rev, 2007; 8:13–17.

Cawley

, Meyerhoefer

. National Bureau of Economic Research. The medical care costs of obesity: An instrumental variables approach. http://www.nber.org/papers/w16467. 2011 July 19.

Avenell

, Broom

, Brown

et al. Systematic review of the long-term effects and economic consequences of treatments for obesity and implications for health improvement. Health Technol Assess, 2004; 8:1–182.

Fujioka

. Benefits of moderate weight loss in patients with type 2 diabetes. Diabetes Obes Metab, 2010; 12:186–194.

Pasanisi

, Contaldo

, de Simone

, Mancini

. Benefits of sustained moderate weight loss in obesity. Nutr Metab Cardiovasc Dis, 2001; 11:401–406.

10.

NHLBI Obesity Education Initiative. The Practical Guide: Identification, Evaluation, and Treatment of Overweight and Obesity in Adults. http://www.nhlbi.nih.gov/guidelines/obesity/prctgd_c.pdf. 2011 July 19.

11.

US Preventive Services Task Force. Screening for Obesity in Adults: Recommendations and Rationale. http://www.uspreventiveservicestaskforce.org/3rduspstf/obesity/obesrr.htm. 2011 July 19.

12.

Abid

, Galuska

, Khan

, Gillespie

, Ford

, Serdula

. Are healthcare professionals advising obese patients to lose weight? A trend analysis. Med Gen Med, 2005; 7:10.

13.

Jackson

, Doescher

, Saver

, Hart

. Trends in professional advice to lose weight among obese adults, 1994 to 2000. J Gen Intern Med, 2005; 20:814–818.

14.

Huang

, Yu

, Marin

, Brock

, Carden

, Davis

. Physicians' weight loss counseling in two public hospital primary care clinics. Acad Med, 2004; 79:156–161.

15.

Greenwood

. The complexity of weight loss counseling. J Am Board Fam Med, 2009; 22:113–114.

16.

Pagoto

, Ockene

. Windows of opportunity for smoking and weight loss counseling. Arch Intern Med, 2009; 169:217–218.

17.

Kushner

, Weinsier

. Evaluation of the obese patient. Practical considerations. Med Clin North Am, 2000; 84:387–399.

18.

, Xiao

, Stafford

. Underdiagnosis of obesity in adults in US outpatient settings. Arch Intern Med, 2009; 169:313–314.

19.

Stafford

, Farhat

, Misra

, Schoenfeld

. National patterns of physician activities related to obesity management. Arch Fam Med, 2000; 9:631–638.

20.

Bleich

, Pickett-Blakely

, Cooper

. Physician practice patterns of obesity diagnosis and weight-related counseling. Patient Educ Couns, 2011; 82:123–129.

21.

Andersen

. Revisiting the behavioral model and access to medical care: Does it matter? J Health Soc Behav, 1995; 36:1–10.

22.

Kushner

. Barriers to providing nutrition counseling by physicians: A survey of primary care practitioners. Prev Med, 1995; 24:546–552.

23.

Dorsey

, Mauldon

, Magraw

, Valka

, Yu

, Krumholz

. Applying practice recommendations for the prevention and treatment of obesity in children and adolescents. Clin Pediatr, 2010; 49:137–145.

24.

Hebl

, Xu

. Weighing the care: Physicians' reactions to the size of a patient. Int J Obes Relat Metab Disord, 2001; 25:1246–1252.

25.

Sussman

, Williams

, Leverence

, Gloyd

Jr , Crabtree

. The art and complexity of primary care clinicians' preventive counseling decisions: Obesity as a case study. Ann Fam Med, 2006; 4:327–333.

26.

Hash

, Munna

, Vogel

, Bason

. Does physician weight affect perception of health advice? Prev Med, 2003; 36:41–44.

27.

Laws

, Jayasinghe

, Harris

, Williams

, Powell Davies

, Kemp

. Explaining the variation in the management of lifestyle risk factors in primary health care: A multilevel cross sectional study. BMC Public Health, 2009; 9:165.

28.

Bardia

, Holtan

, Slezak

, Thompson

. Diagnosis of obesity by primary care physicians and impact on obesity management. Mayo Clin Proc, 2007; 82:927–932.

29.

Waring

, Roberts

, Parker

, Eaton

. Documentation and management of overweight and obesity in primary care. J Am Board Fam Med, 2009; 22:544–552.

30.

Tang

, Kushner

, Thompson

, Baker

. Physician counseling of young adults with rapid weight gain: A retrospective cohort study. BMC Fam Pract, 2010; 11:31.

31.

Bordowitz

, Morland

, Reich

. The use of an electronic medical record to improve documentation and treatment of obesity. Fam Med, 2007; 39:274–279.

32.

Jia

, Lubetkin

. The impact of obesity on health-related quality-of-life in the general adult US population. J Public Health, 2005; 27:156–164.

33.

Jia

, Lubetkin

. Trends in quality-adjusted life-years lost contributed by smoking and obesity. Am J Prev Med, 2010; 38:138–144.

34.

Dixon

. The effect of obesity on health outcomes. Mol Cell Endocrinol, 2010; 316:104–108.

35.

Freedman

, Sigurdson

, Rajaraman

, Doody

, Linet

, Ron

. The mortality risk of smoking and obesity combined. Am J Prev Med, 2006; 31:355–362.

36.

Koster

, Leitzmann

, Schatzkin

et al. The combined relations of adiposity and smoking on mortality. Am J Clin Nutr, 2008; 88:1206–1212.

37.

Stewart

, Cutler

, Rosen

. Forecasting the effects of obesity and smoking on U.S. life expectancy. N Engl J Med, 2009; 361:2252–2260.

38.

Neovius

, Sundstrom

, Rasmussen

. Combined effects of overweight and smoking in late adolescence on subsequent mortality: Nationwide cohort study. BMJ, 2009; 338:b496.